Linux-Development-Sys Digest #704, Volume #6 Wed, 12 May 99 15:14:11 EDT
Contents:
Re: what means cli() and sti()? (Lew Pitcher)
RH6.0 package dependencies (Was : Glibc rant) (Marc Lefranc)
Re: Where can I get reference about linux device driver? (Curt Corum)
Re: Glibc rant (Philipp Thomas)
Linking under SuSe 6.0 (Daniel Sawitzki)
Re: Reliable (!) nic for 2.2 kernel? (Johan Kullstam)
Re: rpm and Data Typ 9 (Gerald Willmann)
Re: Reliable (!) nic for 2.2 kernel? (Leslie Mikesell)
Upgrading from libc4 on a RiscPC? (Matthew Vernon)
glibc-2.0.7 to glibc-2.1.1 (Brent Corbin)
Re: You can now use Winmodems in Linux!!!!!!! (Philip Boucherat)
Re: 2.2.6 doesn't boot on a 386? (Matthew Vernon)
Re: Linux disk defragmenter (Stefaan A Eeckels)
Re: Reliable (!) nic for 2.2 kernel? (bryan)
I want to learn to make drivers (Enrique Cespedes)
Re: glibc-2.0.7 to glibc-2.1.1 ("G. Sumner Hayes")
What is the status of uvfat? (Dick Repasky)
Re: Reliable (!) nic for 2.2 kernel? (bryan)
Re: Glibc rant ("G. Sumner Hayes")
Re: RH6.0 package dependencies (Was : Glibc rant) (David Fox)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: what means cli() and sti()?
Reply-To: [EMAIL PROTECTED]
Date: Wed, 12 May 1999 11:58:33 GMT
cli clear interrupts (disables interrupt handlers)
sti set interrupts (enables interrupt handlers)
typically used in kernel code at points where interrupts
would either cause unintended side-effects, or where an
atomic action must be taken.
On 12 May 1999 00:26:17 -0400, Keith Wright <[EMAIL PROTECTED]> wrote:
>
>But these are not system calls, they are machine instructions.
>~$man cli
>No manual entry for cli
>
>Len Huppe <[EMAIL PROTECTED]> writes:
>
>> You can find information on system calls with man on most Unix systems
>> including Linux.. Run a 'man man' and you'll see that there's a whole
>> section on system calls.
>>
>> good luck
>>
>> lckun wrote:
>>
>> > What means the functions cli() and sti() in the source sched.c of
>> > kernel?
>> >
>
>--
> -- Keith Wright <[EMAIL PROTECTED]>
>
>Programmer in Chief, Free Computer Shop <http://www.free-comp-shop.com>
> --- Food, Shelter, Source code. ---
Lew Pitcher
System Consultant, Integration Solutions Architecture
Toronto Dominion Bank
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
From: Marc Lefranc <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: RH6.0 package dependencies (Was : Glibc rant)
Date: 12 May 1999 14:44:59 +0200
There is something I quite don't understand in RH6.0 regarding the
upgrade to glibc-2.1.
Most packages in RH6.0 depend on libc.so.6(GLIBC_2.0) and
libm.so.6(GLIBC_2.0), as well as on their 2.1 counterparts:
1009 # rpm -q --whatrequires "libc.so.6(GLIBC_2.0)" | wc
362 362 5797
1010 # rpm -qa | wc
523 523 8834
1016 # rpm -q --whatrequires "libc.so.6(GLIBC_2.0)" | grep XFree86
XFree86-Xnest-3.3.3.1-49
XFree86-SVGA-3.3.3.1-49
XFree86-VGA16-3.3.3.1-49
XFree86-XF86Setup-3.3.3.1-49
XFree86-libs-3.3.3.1-49
XFree86-xfs-3.3.3.1-49
XFree86-3.3.3.1-49
XFree86-devel-3.3.3.1-49
1017 # rpm -q --whatrequires "libc.so.6(GLIBC_2.1)" | grep XFree86
XFree86-Xnest-3.3.3.1-49
XFree86-SVGA-3.3.3.1-49
XFree86-VGA16-3.3.3.1-49
XFree86-XF86Setup-3.3.3.1-49
XFree86-libs-3.3.3.1-49
XFree86-xfs-3.3.3.1-49
XFree86-3.3.3.1-49
XFree86-devel-3.3.3.1-49
I naively thought they should only depend on libc.so.6(GLIBC_2.1) only?
Any comment ?
TIA,
Marc
P.S. I have also observed that a rpm package freshly built from one of
my codes has this dependency on both libc.so.6(GLIBC_2.1) and
libc.so.6(GLIBC_2.0). Perhaps, there is a problem with the compiler ?
--
_____________________________________________________________
Marc Lefranc, Charge de Recherche au CNRS
Laboratoire de Physique des Lasers, Atomes, Molecules
Bat P5, UFR de Physique
Universite des Sciences et Technologies de Lille
F-59655 Villeneuve d'Ascq CEDEX (FRANCE)
e-mail: [EMAIL PROTECTED] ; FAX : +33 (0)3 20 33 70 20
_____________________________________________________________
------------------------------
From: Curt Corum <[EMAIL PROTECTED]>
Subject: Re: Where can I get reference about linux device driver?
Date: Tue, 11 May 1999 22:01:50 -0700
Try this book (amazon link below)
Linux Device Drivers
by Alessandro Rubini, Andy Oram (Editor)
http://www.amazon.com/exec/obidos/ASIN/1565922921/n/qid=926485129/sr=2-1/002-5645852-5783445
For some online info:
http://www.redhat.com/~johnsonm/devices.html
-Curt
������ wrote:
> I am developing some PCI board devcie driver for linux. But I have few
> information
>
> about linux device driver.Where can I get some information?
>
> Thanks for reading, have a nice day!
------------------------------
From: [EMAIL PROTECTED] (Philipp Thomas)
Subject: Re: Glibc rant
Date: Wed, 12 May 1999 13:10:53 GMT
On Mon, 10 May 1999 10:25:22 -0400, "Gerry S. Hayes" <[EMAIL PROTECTED]>
wrote:
>People like StarOffice may not have even known that when they built it
>on Red Hat 5.0 (or whatever platform) that they were using a beta glibc.
But they have had to be aware that they where calling an internal function
(clearly visible by the name with leading underscores), which is a definitive
nono regardless of whether its a development or release version.
And recalling an article in the german magazine c't about the initial port of
StarOffice to Linux shows that these are developers that know what they're
doing (like writing perl scripts to fix object code g++ miscompiled).
Philipp
--
You have moved your mouse. Windows must be rebooted for the
changes to take effect.
------------------------------
From: Daniel Sawitzki <[EMAIL PROTECTED]>
Subject: Linking under SuSe 6.0
Date: Wed, 12 May 1999 17:04:38 +0200
Hello
I've got the following problem:
I wanted to compile something under SuSe 6.0. I found out that the
Shared Libs in the /usr/X11R6/lib and /usr/X386/lib (and some other dir)
were not found. Of course I checkes it out with ldconfig -v and ldconfig
-p. But there all bibs are linked correctly. When calling gcc test.c
-lvga, this works fine. But for example gcc test.c -lpng isn't found. I
made no radical changes to my System. Once I killed all files in
/usr/local/lib, but there were only lesstif libs. This (Lesstif) was in
fact the only lib i compiled myself on the system. The libs are there in
the /usr/X11R6/lib dir. What can be the problem?
Thanks
Daniel Sawitzki
------------------------------
From: Johan Kullstam <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: Reliable (!) nic for 2.2 kernel?
Date: 12 May 1999 10:35:28 -0400
[EMAIL PROTECTED] (Leslie Mikesell) writes:
> In article <[EMAIL PROTECTED]>,
> Johan Kullstam <[EMAIL PROTECTED]> wrote:
>
> >> : there's a 10/100 switch which i have set to 100.
> >>
> >> sounds like a single mode hub (choose 1: 10 or 100, and everyone on
> >> the hub MUST speak that speed. right?)
> >
> >yes. exactly.
> >
> [...]
> >i only have 2 computers. they are both in 100/full duplex mode.
>
> If you are using a hub you have to use half duplex. Full duplex
> only works with switches.
ok thanks! i didn't know that.
does full duplex mode work with a direct cross-over cable?
--
johan kullstam
------------------------------
From: Gerald Willmann <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: rpm and Data Typ 9
Date: Wed, 12 May 1999 07:38:49 -0700
On Wed, 12 May 1999, Folkert Meeuw wrote:
> Hi Dear Friendly Readers,
> what is Data Typ 9 and why is it not supported by rpm ?
are you running some old RH4? Upgrade to the latest rpm.rpm from the RH4.2
updates and all should be fine.
> MfG Folkert Meeuw
if you don't want to spell this out cut it short - especially in an
English speaking ng.
Gerald
------------------------------
From: [EMAIL PROTECTED] (Leslie Mikesell)
Crossposted-To: comp.os.linux.networking
Subject: Re: Reliable (!) nic for 2.2 kernel?
Date: 11 May 1999 23:38:41 -0500
In article <[EMAIL PROTECTED]>,
Johan Kullstam <[EMAIL PROTECTED]> wrote:
>> : there's a 10/100 switch which i have set to 100.
>>
>> sounds like a single mode hub (choose 1: 10 or 100, and everyone on
>> the hub MUST speak that speed. right?)
>
>yes. exactly.
>
[...]
>i only have 2 computers. they are both in 100/full duplex mode.
If you are using a hub you have to use half duplex. Full duplex
only works with switches.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: Matthew Vernon <[EMAIL PROTECTED]>
Subject: Upgrading from libc4 on a RiscPC?
Date: 12 May 1999 16:34:50 +0100
Hi,
A friend of mine has an Arm machine (normally runs RiscOS),
though the processor is not the same as that of the Sidewinder. His
version of linux, however is still a libc4 system. Is it possible for
him to natively upgrade by compiling, or would gcc on my i386 box be
able to cross-compile the relavent libraries (and gcc) for him?
Thanks,
Matthew
--
Elen Sila Lumenn' Omentielvo
Matthew Vernon, Steward of the Cambridge Tolkien Society
Selwyn College Computer support. Email: [EMAIL PROTECTED]
http://pick.sel.cam.ac.uk http://www.cam.ac.uk/CambUniv/Societies/tolkien/
------------------------------
From: [EMAIL PROTECTED] (Brent Corbin)
Subject: glibc-2.0.7 to glibc-2.1.1
Date: 12 May 1999 15:36:25 GMT
Well... I confess, some time ago I managed to convert my entire
system over to glibc-2.0.7 without paying (enough) attention to
all the various warnings to consider it unstable... (hey - it was
stable enough for most of the distributions 8*)
Now it's time to move towards glibc-2.1.1 ---- let's say I was
to install the precompiled rpm's from RedHat - can anyone tell
me what would break?
I guess I'm asking whether glibc-2.1.1 is compatible enough that
I maintain the most important aspects of functionality on my
system (bash, telnet, ftp, make, ls...) --- if it is, then there's
hope of fixing other compatibility problems piece-by-piece -
if not... <mumble mumble>
Thanks... //Brent
------------------------------
From: Philip Boucherat <[EMAIL PROTECTED]>
Subject: Re: You can now use Winmodems in Linux!!!!!!!
Date: Wed, 12 May 1999 16:50:38 +0100
What is a Winmodem anyway, and why would I want to use it?
--
Philip Boucherat
------------------------------
From: Matthew Vernon <[EMAIL PROTECTED]>
Subject: Re: 2.2.6 doesn't boot on a 386?
Date: 12 May 1999 16:29:30 +0100
Olav Woelfelschneider <[EMAIL PROTECTED]> writes:
> Tried to boot some kernels on an old 386DX40. This beast should be revived
> as an ISDN/Ethernet gateweay later.
>
> 2.1.130 boots fine.
>
> 2.2.6 however loads, uncompresses, says "now booting the kernel", puts
> the cursor on the next line and hangs. This happens with both zImage and
> bzImage.
FWIW, my webserver runs 2.2.1 fine (compiled on my PII ;) ). Did you
set the target processor type correctly?
Matthew
--
Elen Sila Lumenn' Omentielvo
Matthew Vernon, Steward of the Cambridge Tolkien Society
Selwyn College Computer support. Email: [EMAIL PROTECTED]
http://pick.sel.cam.ac.uk http://www.cam.ac.uk/CambUniv/Societies/tolkien/
------------------------------
From: [EMAIL PROTECTED] (Stefaan A Eeckels)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Linux disk defragmenter
Date: 12 May 1999 13:30:47 GMT
In article <7hbnn7$3u4$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (M Sweger) writes:
> Some questions I have are,
>
> a). is fsck for different linux filesystems equivalent to or serve the
> same purpose as MSofts scandisk. That is to repair the filesystem?
It is indeed used to verify the integrity of the file system after
an uncontrolled shutdown. The ext2 data structures are more sophisticated
than those of FAT, and there's more redundancy, hence the repair operation
tends to be more transparent than on FAT.
> If a run Msofts scandisk on a UMSMDOS filesystem to verify the
> Linux files mapped to msdos ones, it starts finding cross-links
> and lost clusters. If I answer yes to everyone, then I have my
> Linux UMSDOS filesystem butchered to pieces and unusable.
I don't use UMSDOS - way too kludgy, and FAT is a really grotty file
system. I don't think that UMSDOS relies on creating a corrupted file
system though.
> b). what would be equivalent to Msoft defrag or Norton utilities defrag
> for linux on an ext2 filesystem vs. an UMSDOS filesystem.
There is a defragger for ext2, but on the whole ext2 doesn't need
to be defragmented. FAT, VFAT and NTFS suffer badly from fragmentation,
also through the lack of structuring the file system into related
file sets on Linux (/ for all the boot-level stuff, /usr, /opt,
/home, /var etc). Cramming all the files in a single volume (C:)
is a rather silly idea. I've a Linux server that has been running
since 1996, and even /homes has only minimal fragmentation.
> If I run Msofts defrag on a UMSDOS filesystem(without booting int linux)
> It gets about 25-50% of the way thru, then complains it is out of
> memory. I have 128 megs of memory for Msofts defrag program.
How big is the file system? I've seen the same behaviour on a
4Gb volume, and had to run defrag about 8 times to defragment the
complete disk.
--
Stefaan
--
PGP key available from PGP key servers (http://www.pgp.net/pgpnet/)
___________________________________________________________________
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away. -- Saint-Exup�ry
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: Reliable (!) nic for 2.2 kernel?
Crossposted-To: comp.os.linux.networking
Date: Wed, 12 May 1999 16:34:30 GMT
In comp.os.linux.networking Johan Kullstam <[EMAIL PROTECTED]> wrote:
: [EMAIL PROTECTED] (Leslie Mikesell) writes:
: > In article <[EMAIL PROTECTED]>,
: > Johan Kullstam <[EMAIL PROTECTED]> wrote:
: >
: > >> : there's a 10/100 switch which i have set to 100.
: > >>
: > >> sounds like a single mode hub (choose 1: 10 or 100, and everyone on
: > >> the hub MUST speak that speed. right?)
: > >
: > >yes. exactly.
: > >
: > [...]
: > >i only have 2 computers. they are both in 100/full duplex mode.
: >
: > If you are using a hub you have to use half duplex. Full duplex
: > only works with switches.
: ok thanks! i didn't know that.
and its not universally true - so don't believe it. it depends on the
implementation. there are no hard and fast rules on network gear
anymore - each vendor tries to differentiate themselves.
: does full duplex mode work with a direct cross-over cable?
of course.
: --
: johan kullstam
--
Bryan
------------------------------
From: Enrique Cespedes <[EMAIL PROTECTED]>
Subject: I want to learn to make drivers
Date: Sun, 09 May 1999 14:17:10 +0200
Hi, I'd like to learn how to make drivers. Here can I found
documentation. There is a site at the wed ? Or a good book ?
Thanks
Enrique Cespedes Sanchez.
------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Subject: Re: glibc-2.0.7 to glibc-2.1.1
Date: Wed, 12 May 1999 12:41:23 -0400
Brent Corbin wrote:
>
> Well... I confess, some time ago I managed to convert my entire
> system over to glibc-2.0.7 without paying (enough) attention to
> all the various warnings to consider it unstable... (hey - it was
> stable enough for most of the distributions 8*)
>
> Now it's time to move towards glibc-2.1.1 ---- let's say I was
> to install the precompiled rpm's from RedHat - can anyone tell
> me what would break?
>
> I guess I'm asking whether glibc-2.1.1 is compatible enough that
> I maintain the most important aspects of functionality on my
> system (bash, telnet, ftp, make, ls...)
Yes. Read the README -- ncurses will break, as will a few other
things. This will make less break (which could make things a bit
less convenient when fixing the rest of the system) and might break
your favorite editor, depending on what it links against (but you
do know how to use ed, the standard text editor, right? And you
have it installed, right? ;) bash, telnet, ftp, make, etc. all ought
to work fine.
So make sure you have ncurses ready to go before you upgrade everything.
For safety's sake, try using LD_PRELOAD to test the new libs after you
build them but before you install them.
You imply you upgraded to glibc-2.0.7(beta C?) by hand; if so, the
glibc 2.0->2.1 move really goes pretty smoothly.
--Sumner
------------------------------
From: [EMAIL PROTECTED] (Dick Repasky)
Subject: What is the status of uvfat?
Date: 12 May 1999 13:54:53 GMT
Reply-To: [EMAIL PROTECTED]
Hi,
What ever happened to uvfat, and what is its status? It is absent
from freshmeat, and all references on the new, anti-improved deja news
are very old.
Thanks,
Dick
--
Remove the underscore from my e-mail address to reply by mail.
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: Reliable (!) nic for 2.2 kernel?
Crossposted-To: comp.os.linux.networking
Date: Wed, 12 May 1999 13:58:19 GMT
In comp.os.linux.networking Leslie Mikesell <[EMAIL PROTECTED]> wrote:
: In article <[EMAIL PROTECTED]>,
: Johan Kullstam <[EMAIL PROTECTED]> wrote:
: >> : there's a 10/100 switch which i have set to 100.
: >>
: >> sounds like a single mode hub (choose 1: 10 or 100, and everyone on
: >> the hub MUST speak that speed. right?)
: >
: >yes. exactly.
: >
: [...]
: >i only have 2 computers. they are both in 100/full duplex mode.
: If you are using a hub you have to use half duplex. Full duplex
: only works with switches.
not always true. depends on the implementation.
I believe the bay net small 10/100 hub will support 100/full and even
10/full on its ports.
--
Bryan
------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss
Subject: Re: Glibc rant
Date: Wed, 12 May 1999 12:53:10 -0400
"Thomas Bushnell, BSG" wrote:
>
> It is not at all unreasonable for shared libraries to occasionally
> have to bump the soname. That does not require "every application to
> be rebuilt", it requires people to have both versions of the library
> on disk.
That's perfectly fine with me. I'll note that I personally have had
no problems with the move from glibc 2.0->2.1; the issue is that there
rather a lot of people complaining on c.o.l.d.s about how the move
seriously breaks backwards compatibility. I'm trying to pinpoint
exactly where the problem is (possibly distributions moving to glibc2.1
without installing 2.0 by default, possibly end users not understanding
why they have two glibcs and erasing one, possibly real glibc problems).
Your implication is that if I have a glibc-2.0 system installed and I
grab glibc-2.1 and do "configure; make; make install" that the default
sonames are properly bumped and old programs will continue to use
glibc-2.0, right? I honestly don't know, as I moved to glibc-2.1 and
egcs at the same time and rebuilt the whole system. If that is the
case, then the question is how to educate users and/or distributions
about the right way to migrate to avoid problems. One possible idea
might be for some of the Linux folks with an interest in the issue to
get a little group together with the sole purpose of communicating with
the glibc folks prior to new releases, finding out the proper way to
update systems, and passing this information along to the various
distribution packagers. Or, if the distributions are doing things
properly already, make sure that they explain clearly to the end-users
why there are two sets of glibc2 libraries (which could be a simple
as a note in the RPM/DPG info for the package and a README stuck in
/lib).
--Sumner
------------------------------
From: d s f o x @ c o g s c i . u c s d . e d u (David Fox)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: RH6.0 package dependencies (Was : Glibc rant)
Date: 12 May 1999 07:12:53 -0700
Marc Lefranc <[EMAIL PROTECTED]> writes:
> There is something I quite don't understand in RH6.0 regarding the
> upgrade to glibc-2.1.
>
> Most packages in RH6.0 depend on libc.so.6(GLIBC_2.0) and
> libm.so.6(GLIBC_2.0), as well as on their 2.1 counterparts:
>
> 1009 # rpm -q --whatrequires "libc.so.6(GLIBC_2.0)" | wc
> 362 362 5797
> 1010 # rpm -qa | wc
> 523 523 8834
>
> 1016 # rpm -q --whatrequires "libc.so.6(GLIBC_2.0)" | grep XFree86
> XFree86-Xnest-3.3.3.1-49
> XFree86-SVGA-3.3.3.1-49
> XFree86-VGA16-3.3.3.1-49
> XFree86-XF86Setup-3.3.3.1-49
> XFree86-libs-3.3.3.1-49
> XFree86-xfs-3.3.3.1-49
> XFree86-3.3.3.1-49
> XFree86-devel-3.3.3.1-49
> 1017 # rpm -q --whatrequires "libc.so.6(GLIBC_2.1)" | grep XFree86
> XFree86-Xnest-3.3.3.1-49
> XFree86-SVGA-3.3.3.1-49
> XFree86-VGA16-3.3.3.1-49
> XFree86-XF86Setup-3.3.3.1-49
> XFree86-libs-3.3.3.1-49
> XFree86-xfs-3.3.3.1-49
> XFree86-3.3.3.1-49
> XFree86-devel-3.3.3.1-49
>
> I naively thought they should only depend on libc.so.6(GLIBC_2.1) only?
>
> Any comment ?
>
> TIA,
> Marc
>
> P.S. I have also observed that a rpm package freshly built from one of
> my codes has this dependency on both libc.so.6(GLIBC_2.1) and
> libc.so.6(GLIBC_2.0). Perhaps, there is a problem with the compiler ?
I would love to know the answer to this, as my own programs link with
dependencies on 2.0 symbols. In fact, a simple test program such as
#include <stdio.h>
main()
{
char *user = getenv("USER");
printf("user: %s\n", user);
}
ends up pulling in GLIBC_2.0 symbols:
U __deregister_frame_info@@GLIBC_2.0
U __libc_start_main@@GLIBC_2.0
U __register_frame_info@@GLIBC_2.0
U getenv@@GLIBC_2.0
U printf@@GLIBC_2.0
This is a problem for me because libraries like ElectricFence that try
to override malloc don't work, because they can't override
malloc@@GLIBC_2.0.
--
David Fox http://hci.ucsd.edu/dsf xoF divaD
UCSD HCI Lab baL ICH DSCU
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
